List your company
ResearchFormerly WizWand

Class-Incremental Learning on CIFAR100 5 tasks

89.4AccuracyJoint
17.723236.331654.9473.5484May 8, 2026May 9, 2026May 10, 2026May 11, 2026May 12, 2026
Updated 5mo ago

Evaluation results

MethodLinks
JointBackbone=CLIP ViT-B/16, Method Category=Baseline2026.0589.4
TRMBackbone=CLIP ViT-B/16, Method Category=Model Merging2026.0583.5
BECAMEYear=2025, Backbone=CLIP ViT-B/16, Method Category=Model Merging2026.0580.9
MagMaxYear=2024, Backbone=CLIP ViT-B/16, Method Category=Model Merging2026.0580.4
TIESYear=2023, Backbone=CLIP ViT-B/16, Method Category=Model Merging2026.0579.8
RAPFYear=2024, Backbone=CLIP ViT-B/16, Method Category=PEFT2026.0579.2
CLG-CBMYear=2025, Backbone=CLIP ViT-B/16, Method Category=PEFT2026.0578.1
DualPromptYear=2023, Backbone=CLIP ViT-B/16, Method Category=PEFT2026.0577.4
CODAPromptYear=2023, Backbone=CLIP ViT-B/16, Method Category=PEFT2026.0577.3
L2PYear=2022, Backbone=CLIP ViT-B/16, Method Category=PEFT2026.0577.2
Model StockYear=2024, Backbone=CLIP ViT-B/16, Method Category=Model Merging2026.0576.5
Upper boundModel Inversion=false, Backbone=ResNet-32, Number of runs=52026.0570.59
Upper boundModel inversion=false, Backbone=ResNet-32, Number of runs=52026.0570.59
EWCYear=2017, Backbone=CLIP ViT-B/16, Method Category=Conventional2026.0569.3
LwFYear=2017, Backbone=CLIP ViT-B/16, Method Category=Conventional2026.0568.2
Zero-shotBackbone=CLIP ViT-B/16, Method Category=Baseline2026.0566.7
REMIXModel Inversion=true, Backbone=ResNet-32, Number of runs=52026.0552.94
REMIXModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0552.94
PMIModel Inversion=true, Backbone=ResNet-32, Number of runs=52026.0552.38
PMIModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0552.38
PMI w/o CFSModel Inversion=true, Backbone=ResNet-32, Number of runs=52026.0552.05
PMI w/o CFSModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0552.05
R-DFCILModel Inversion=true, Backbone=ResNet-32, Number of runs=52026.0549.87
R-DFCILModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0549.87
ABDModel Inversion=true, Backbone=ResNet-32, Number of runs=52026.0548.84
ABDModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0548.84
DCMIModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0541.05
PRAKAModel inversion=false, Backbone=ResNet-32, Number of runs=52026.0537.74
SSREModel inversion=false, Backbone=ResNet-32, Number of runs=52026.0530.39
DeepInversionModel Inversion=true, Backbone=ResNet-32, Number of runs=52026.0520.48
DeepInversionModel inversion=true, Backbone=ResNet-32, Number of runs=52026.0520.48